US7584181B2 - Implicit links search enhancement system and method for search engines using implicit links generated by mining user access patterns - Google Patents
Implicit links search enhancement system and method for search engines using implicit links generated by mining user access patterns Download PDFInfo
- Publication number
- US7584181B2 US7584181B2 US10/676,794 US67679403A US7584181B2 US 7584181 B2 US7584181 B2 US 7584181B2 US 67679403 A US67679403 A US 67679403A US 7584181 B2 US7584181 B2 US 7584181B2
- Authority
- US
- United States
- Prior art keywords
- computer
- implicit
- implicit links
- links
- search
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related, expires
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9538—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10—TECHNICAL SUBJECTS COVERED BY FORMER USPC
- Y10S—TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10S707/00—Data processing: database and file management or data structures
- Y10S707/99931—Database or file accessing
- Y10S707/99933—Query processing, i.e. searching
- Y10S707/99935—Query augmenting and refining, e.g. inexact access
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10—TECHNICAL SUBJECTS COVERED BY FORMER USPC
- Y10S—TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10S707/00—Data processing: database and file management or data structures
- Y10S707/99931—Database or file accessing
- Y10S707/99937—Sorting
Definitions
- the present invention relates in general to computer search engines and more particularly to an implicit links enhancement system and method for search engines that generates implicit links obtained from mining user access logs to provide accurate and efficient local searching of web sites and intranets.
- Search engines are vital for helping a user find specific information in the vast expanse of the World Wide Web (WWW or Web). Because the Web continues to grow at a phenomenal rate, it would be virtually impossible to locate anything on the Web without knowing a specific address if not for search engines.
- a search engine refers to a system that maintains an index structure of a collection of documents to efficiently generate a list of documents that contain specified keywords and ranks the document list according to a relevance measurement.
- Global search engines which are popular and widespread, are used to search the entire Web, while local search engines are used to search web sites and intranets.
- search engines use link analysis to quickly and efficiently search the entire Web. These search engines analyze links to rank web sites (or pages) according to, among other things, the quality and quantity of other sites that are linked to them.
- a link in a hypertext context such as the Web
- the search engine will give the particular web site because more links indicates a higher level of popularity among users.
- PageRank is describe in a paper by L. Page, S. Brin, R. Motwani and T. Winograd entitled “The PageRank citation ranking: bringing order to the Web” in a Technical report, Stanford University Database Group, 1998 and in a paper by S. Brin and L. Page entitled “The anatomy of a large-scale hypertextual web search engine” in Proc. of WWW 7, 107-117, Brisbane, Australia, April 1998.
- each web-page w i has both a hub score h i and an authority score a i .
- the hub score of w i is the sum of all the authority scores of pages that are pointed by w i ; the authority score of w i is the sum of all the hub scores of pages that point to w i , as shown in the following equations.
- PageRank is a core algorithm of the popular Google search engine (http://www.google.com.). PageRank measures the importance of web pages. specifically, PageRank uses the whole linkage graph of the Web to compute universal query-independent rank value for each page.
- a users' browsing model is modeled as a random surfing model. This model assumes that a user either follows a link from a current page or jumps to a random page in the graph.
- the PageRank of a page w i then is computed by the following equation:
- PR ⁇ ( w i ) ⁇ n + ( 1 - ⁇ ) ⁇ ⁇ l j , i ⁇ E ⁇ PR ⁇ ( w j ) / outdegree ⁇ ( w j )
- ⁇ is a dampening factor, which is usually set between 0.1 and 0.2
- n is the number of nodes in G
- out-degree(w j ) is the number of the edges leaving page w j (i.e., the number of hyperlinks on page w j ).
- the PageRank can be computed by an iterative algorithm and corresponds to the primary eigenvector of a matrix derived from adjacency matrix of the available portion of the Web.
- a web site can be thought of as a closed space on the web where data and information are available to a user.
- web sites include enterprise portals (allowing document access and product information), server providers (including access to news and magazines), education institutions providing online courses and document access, and user groups, to name a few.
- enterprise portals allowing document access and product information
- server providers including access to news and magazines
- education institutions providing online courses and document access
- user groups to name a few.
- global search engines are also impractical for local searching because the link structure of a web site and intranet is different from the Web. In the closed sub-space of a web site or intranet local search engines must used.
- link analysis uses explicit links to a certain site to determine the ranking of the site. While this recommendation assumption is generally correct for the Web, it is commonly invalid for a Web site or intranet. In general, this is because there are relatively few explicit links and the links are created by a small number of authors whose purpose is to organize the contents into a hierarchical structure. Thus, in general the authority of pages is not captured correctly by link analysis.
- DirectHit http://www.directhit.com
- DirectHit's site ranking system which is based on the concepts of “click popularity” and “stickiness,” is currently used by Lycos, Hotbot, MSN, Infospace, About.com and several other search engines. The underlying assumption is that the more a web-page is visited, the higher it is ranked according to particular queries.
- These usage-based search engines have restrictions. In particular, one problem is that the technique requires large amounts of user logs and only works for some popular queries. Another problem is that it is easy to fall into a quick positive feedback loop when access to a popular resource leads to its higher rank. This in turn leads to an even higher number accesses to it.
- the invention disclosed herein includes an implicit links search enhancement system and method that generates implicit links obtained from mining user access logs to facilitate enhanced local searching of Web sub-space (such as web sites and intranets).
- the implicit links search enhancement system and method augments traditional link analysis search engines popular for global Web searches and makes them available for local searching of Web sub-space.
- the implicit links search enhancement system and method extracts implicit links in addition to explicit links and filters out unimportant links to achieve improved search results.
- the initial search results obtained with a tradition link analysis search engine then are updated based on the information provided by the implicit links search enhancement system and method.
- the implicit links search enhancement method includes generating implicit links from a user access log.
- the implicit links are implicit recommendation links. All probably implicit links then are extracted from the user access log using a two-item sequential pattern mining technique. This technique includes using a gliding window to find ordered pairs of implicit links or pages.
- An implicit links graph is constructed using the extracted implicit links. Two-item sequential patterns also are generated from the implicit links and are used to update the implicit links graph. Updated rankings of the search results are made using the updated implicit links graph and a modified implicit links analysis.
- the user access log is pre-processed.
- This pre-processing includes data cleaning, session identification, and consecutive repetition elimination. Data cleaning is performed by filtering out any access entries for embedded objects, such as images and scripts. Browsing sessions are identified by the Internet protocol (IP) address and assumes consecutive accesses from the same IP address during a time interval are from the same user. Consecutive repetition elimination removes IP addresses whose page hits count exceeds some threshold.
- IP Internet protocol
- Consecutive repetition elimination removes IP addresses whose page hits count exceeds some threshold.
- the user access log is segmented into individual browsing sessions. Each browsing session is identified by its user identification and pages in a browsing path ordered by timestamp. The ordered pairs are generated from the segmented user access log. First, a gliding window size is defined. Next, the gliding window is applied to each individual browsing session along the browsing path to generate all possible ordered pairs and their probabilities.
- the ordered pairs are filtered to remove unnecessary links.
- a frequency for each of the ordered pairs is determined.
- a minimum support threshold is defined and applied to the frequency of each of the ordered pairs. If a frequency is below the minimum support threshold, the associated ordered pair is discarded. Otherwise, the ordered pair is kept and used to update the implicit links graph.
- a modified links analysis technique is used to re-rank initial search results.
- the modified links analysis technique uses the updated implicit links graph, a modified re-ranking formula, and at least one of two re-ranking techniques.
- the modified re-ranking formula is a re-ranking formula from PageRank but having novel modifications. One of these modifications is that the traditional PageRank only uses 0 or 1 values for each entry in the adjacency matrix, representing the existence of a hyperlink, while the modified re-ranking formula accommodates any floating point values between 0 and 1.
- the modified links analysis technique uses at least one of two re-ranking techniques: (a) an order-based re-ranking technique; and (b) a score-based re-ranking technique.
- the order-based re-ranking technique is preferred.
- the order-based re-ranking technique uses is based on the rank order of pages.
- the order-based technique is a linear combination of a position of a page in two lists, where one list is sorted by similarity and the other list is sorted by PageRank values.
- the score-based technique uses a linear combination of a content-based similarity score and a PageRank value of all web pages.
- the implicit links search enhancement system is designed to work in unison with a search engine to provide improved search results.
- the system includes a user access log pre-processing module, which performs pre-processing of the user access log, and a user access log segmentation module, which segments the pre-processed log into individual browsing sessions.
- the system also includes an ordered pairs generator and a filter module.
- the ordered pairs generator generates all possible ordered pairs of implicit links and pages from each of the individual browsing sessions.
- the filter module filters the extracted ordered pairs to cull any infrequently occurring links and make the search results re-ranking more accurate.
- the implicit links search enhancement system further includes an updated module, which updates an implicit links graph using the filtered ordered pairs, and a re-ranking module.
- the re-ranking module uses the updated implicit links graph, a modified re-ranking formula, and at least one re-ranking technique to re-rank search results from a search engine into an improved search result.
- FIG. 1 is a block diagram illustrating a general overview of an exemplary implementation of the implicit links search enhancement system and method disclosed herein.
- FIG. 2 illustrates an example of a suitable computing system environment in which the implicit links search enhancement system and method shown in FIG. 1 may be implemented.
- FIG. 3 is a block diagram illustrating the details of an exemplary implementation of the implicit links search enhancement system shown in FIG. 1 .
- FIG. 4 is a general flow diagram illustrating the general operation of the implicit links search enhancement method of the implicit links search enhancement system shown in FIGS. 1 and 3 .
- FIG. 5 is a detailed flow diagram illustrating the operation of the implicit links search enhancement method shown in FIG. 4 and used in the implicit link search enhancement system 100 shown in FIGS. 1 and 3 .
- FIG. 6 is a detailed flow diagram illustrating the operation of the user access log pre-processing module shown in FIG. 3 .
- FIG. 7 is a detailed flow diagram illustrating the operation of the user access log segmentation module shown in FIG. 3 .
- FIG. 8 is a detailed flow diagram illustrating the operation of the ordered pairs generator shown in FIG. 3 .
- FIG. 9 is a detailed flow diagram illustrating the operation of the filter module shown in FIG. 3 .
- FIG. 10 is a detailed flow diagram illustrating the operation of the re-ranking module shown in FIG. 3 .
- FIG. 11 illustrates the precision of page prediction by implicit links in a working example.
- FIG. 12 is a bar graph illustrating the precision and authority of different ranking methods.
- FIG. 13 illustrates the convergence curves of different ranking models.
- FIGS. 14A and 14B illustrate the search precision and implicit link number with different minimum support thresholds.
- FIG. 15 illustrates the impact of different window sizes on search precision.
- FIG. 16 illustrates an interval distribution of implicit links.
- FIG. 17 illustrates the precision of different weighting methods.
- FIG. 18 illustrates the precision of various re-ranking methods.
- a hyperlink in page X pointed to page Y stands for the recommendation of page Y by the author of page X.
- the recommendation assumption is generally correct because hyperlinks encode a considerable amount of authors' judgment.
- some hyperlinks are created not for the recommendation purpose, but their influence could be filtered or reduced to an ignorable level.
- the recommendation assumption commonly is invalid in the case of a small web.
- the majority of hyperlinks in a small web are more “regular” than that in the global Web.
- Most links are from a parent node to children nodes, between sibling nodes, or from leaves to the root (e.g. “Back to Home”).
- the reason is primarily because hyperlinks in a small web are created by a small number of authors.
- the purpose of the hyperlinks is usually to organize the content into a hierarchical or linear structure.
- the in-degree measure does not reflect the authority of pages, making the existing link analysis algorithms not suitable for small web search.
- hyperlinks could be divided into navigational links and recommendation links.
- the latter is useful for link analysis to enhance search.
- only filtering out navigational links from all hyperlinks is inadequate because the remaining recommendation links are incomplete.
- implicit recommendation links hereafter called “implicit links” for short
- the implicit links search enhancement system and method described herein augments conventional search engines to make them more efficient and accurate.
- the implicit links search enhancement system and method includes constructing implicit links by mining users' access patterns and then using a modified link analysis algorithm to re-rank search results obtained from traditional search engines. Experimental results obtained in a working example illustrate that the implicit links search enhancement system and method effectively improves search performance of existing search engines.
- FIG. 1 is a block diagram illustrating a general overview of an exemplary implementation of the implicit links search enhancement system and method disclosed herein.
- the implicit links search enhancement system 100 typically is implemented in a computing environment 110 .
- This computing environment 110 which is described in detail below, includes computing devices (not shown).
- the implicit links search enhancement system 100 augments the search results obtained by a traditional search engine (such as a site search engine 120 ) based on an implicit link analysis.
- a user sends a user query 130 to the site search engine 120 .
- the site may be a web site or an intranet.
- the site search engine 120 obtains pages 140 (such as web pages) and indexes those pages (box 150 ).
- the inverted index 160 is obtained by the site search engine 120 .
- the site search engine 120 obtains and ranks initial search results.
- the implicit links search enhancement system 100 obtains data from a user access log 170 and performs an implicit link analysis on the log 170 . This analysis is described in detail below.
- the implicit links search enhancement system 100 outputs page rankings 180 based on the analysis performed by the implicit links search engine 100 .
- the site search engine 120 uses these page rankings to update the initial search results and output updated search results 190 to the user in response to a query.
- the implicit links search enhancement system and method disclosed herein is designed to operate in a computing environment.
- the following discussion is intended to provide a brief, general description of a suitable computing environment in which the implicit links search enhancement system and method may be implemented.
- FIG. 2 illustrates an example of a suitable computing system environment 200 in which the implicit links search enhancement system and method may be implemented.
- the computing system environment 200 is only one example of a suitable computing environment and is not intended to suggest any limitation as to the scope of use or functionality of the invention. Neither should the computing environment 200 be interpreted as having any dependency or requirement relating to any one or combination of components illustrated in the exemplary operating environment 200 .
- the implicit links search enhancement system and method is operational with numerous other general purpose or special purpose computing system environments or configurations.
- Examples of well known computing systems, environments, and/or configurations that may be suitable for use with the implicit links search engine and method include, but are not limited to, personal computers, server computers, hand-held, laptop or mobile computer or communications devices such as cell phones and PDA's, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.
- the implicit links search enhancement system and method may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer.
- program modules include routines, programs, objects, components, data structures, etc., that perform particular tasks or implement particular abstract data types.
- the implicit links search enhancement system and method may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
- an exemplary system for implementing the implicit links search enhancement system and method includes a general-purpose computing device in the form of a computer 210 .
- Components of the computer 210 may include, but are not limited to, a processing unit 220 , a system memory 230 , and a system bus 221 that couples various system components including the system memory to the processing unit 220 .
- the system bus 221 may be any of several types of bus structures including a memory bus or memory controller, a peripheral bus, and a local bus using any of a variety of bus architectures.
- bus architectures include Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, Enhanced ISA (EISA) bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus also known as Mezzanine bus.
- the computer 210 typically includes a variety of computer readable media.
- Computer readable media can be any available media that can be accessed by the computer 210 and includes both volatile and nonvolatile media, removable and non-removable media.
- Computer readable media may comprise computer storage media and communication media.
- Computer storage media includes volatile and nonvolatile removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data.
- Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by the computer 210 .
- Communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media.
- modulated data signal means a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal.
- communication media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, RF, infrared and other wireless media. Combinations of any of the above should also be included within the scope of computer readable media.
- the system memory 230 includes computer storage media in the form of volatile and/or nonvolatile memory such as read only memory (ROM) 231 and random access memory (RAM) 232 .
- ROM read only memory
- RAM random access memory
- BIOS basic input/output system 233
- RAM 232 typically contains data and/or program modules that are immediately accessible to and/or presently being operated on by processing unit 220 .
- FIG. 2 illustrates operating system 234 , application programs 235 , other program modules 236 , and program data 237 .
- the computer 210 may also include other removable/non-removable, volatile/nonvolatile computer storage media.
- FIG. 2 illustrates a hard disk drive 241 that reads from or writes to non-removable, nonvolatile magnetic media, a magnetic disk drive 251 that reads from or writes to a removable, nonvolatile magnetic disk 252 , and an optical disk drive 255 that reads from or writes to a removable, nonvolatile optical disk 256 such as a CD ROM or other optical media.
- removable/non-removable, volatile/nonvolatile computer storage media that can be used in the exemplary operating environment include, but are not limited to, magnetic tape cassettes, flash memory cards, digital versatile disks, digital video tape, solid state RAM, solid state ROM, and the like.
- the hard disk drive 241 is typically connected to the system bus 221 through a non-removable memory interface such as interface 240 , and magnetic disk drive 251 and optical disk drive 255 are typically connected to the system bus 221 by a removable memory interface, such as interface 250 .
- hard disk drive 241 is illustrated as storing operating system 244 , application programs 245 , other program modules 246 , and program data 247 . Note that these components can either be the same as or different from operating system 234 , application programs 235 , other program modules 236 , and program data 237 . Operating system 244 , application programs 245 , other program modules 246 , and program data 247 are given different numbers here to illustrate that, at a minimum, they are different copies.
- a user may enter commands and information into the computer 210 through input devices such as a keyboard 262 and pointing device 261 , commonly referred to as a mouse, trackball or touch pad.
- Other input devices may include a microphone, joystick, game pad, satellite dish, scanner, radio receiver, or a television or broadcast video receiver, or the like. These and other input devices are often connected to the processing unit 220 through a user input interface 260 that is coupled to the system bus 221 , but may be connected by other interface and bus structures, such as, for example, a parallel port, game port or a universal serial bus (USB).
- a monitor 291 or other type of display device is also connected to the system bus 221 via an interface, such as a video interface 290 .
- computers may also include other peripheral output devices such as speakers 297 and printer 296 , which may be connected through an output peripheral interface 295 .
- the computer 210 may operate in a networked environment using logical connections to one or more remote computers, such as a remote computer 280 .
- the remote computer 280 may be a personal computer, a server, a router, a network PC, a peer device or other common network node, and typically includes many or all of the elements described above relative to the computer 210 , although only a memory storage device 281 has been illustrated in FIG. 2 .
- the logical connections depicted in FIG. 2 include a local area network (LAN) 271 and a wide area network (WAN) 273 , but may also include other networks.
- LAN local area network
- WAN wide area network
- Such networking environments are commonplace in offices, enterprise-wide computer networks, intranets and the Internet.
- the computer 210 When used in a LAN networking environment, the computer 210 is connected to the LAN 271 through a network interface or adapter 270 .
- the computer 210 When used in a WAN networking environment, the computer 210 typically includes a modem 272 or other means for establishing communications over the WAN 273 , such as the Internet.
- the modem 272 which may be internal or external, may be connected to the system bus 221 via the user input interface 260 , or other appropriate mechanism.
- program modules depicted relative to the computer 210 may be stored in the remote memory storage device.
- FIG. 2 illustrates remote application programs 285 as residing on memory device 281 . It will be appreciated that the network connections shown are exemplary and other means of establishing a communications link between the computers may be used.
- FIG. 3 is a block diagram illustrating the details of an exemplary implementation of the implicit links search enhancement system 100 shown in FIG. 1 .
- the implicit links search enhancement system 100 obtains data from the user access log 170 and outputs page rankings based on an implicit links analysis 180 .
- the implicit links search enhancement system 100 includes a number of modules. The function of these modules is described in detail below.
- the modules located in the implicit links search enhancement system 100 include a user access log preprocessing module 300 and a user access log segmentation module 310 .
- the user access log preprocessing module 300 preprocesses the user access log 170 such that the data is cleaned and users are identified.
- the preprocessed data is input for the user access log segmentation module 310 , which segments the data into individual browsing sessions.
- the implicit links search enhancement system 100 also includes an ordered pairs generator 320 and a filter module 330 .
- the ordered pairs generator 320 generates all possible ordered pairs from each of the individual browsing sessions.
- the implicit links search enhancement system 100 also includes an update module 340 and a re-ranking module 350 .
- the remaining ordered pairs from the filter module 330 are input to the update module 340 where the pairs are used to update an implicit link graph.
- the graph is used by the re-ranking module 350 to re-rank the search results (including pages).
- the output from the implicit links search enhancement system 100 are the updated page rankings 180 .
- FIG. 4 is a general flow diagram illustrating the general operation of the implicit links search enhancement method of the implicit links search enhancement system 100 shown in FIGS. 1 and 3 .
- the method begins by segmenting a user access log into a plurality of different browsing sessions (box 400 ).
- implicit links are extracted from the sessions (box 410 ).
- the implicit links are extracted using a two-item sequential pattern mining technique. As explained below, this mining technique uses a gliding window to move over each path in the user access log and generate all ordered pairs.
- An implicit links graph is generated using the extracted implicit links (box 420 ). As discussed below, this implicit links graph is used in place of an explicit link graph used in conventional link analysis techniques. Based on the implicit link graph, a generative model for a user access log can be defined. Given the user access log, this generative model is used to estimate parameters for the log, including the implicit links and their probabilities. Moreover, two-item sequential patterns generated from this mining technique above can be used to update the implicit link graph. Finally, page rankings are computed using the implicit links graph (box 430 ).
- FIG. 5 is a detailed flow diagram illustrating the operation of the implicit links search enhancement method shown in FIG. 4 and used in the implicit link search enhancement system 100 shown in FIGS. 1 and 3 .
- the implicit links search enhancement method begins by pre-processing a user access log (box 500 ). This pre-processing includes cleaning, identification and elimination of redundancies of data in the user access log. Next, the log is segmented into individual browsing sessions (box 510 ). Each browsing session includes a user identification and pages visited in chronological order. Ordered pairs of pages then are generated from the segmented log (box 520 ).
- the ordered pairs of pages then are filtered to remove any pairs that are infrequently occurring (box 530 ). As explained in detail below, this filtering is performed using a minimum support threshold. This generates two-item sequential patterns, which are used to update an implicit link graph (box 540 ).
- the search results are re-ranked (box 550 ).
- the modified link analysis technique includes a modified re-ranking formula and at least one of two types of re-ranking techniques.
- I i,j ⁇ E is used to denote that there exists a link between the page w i and w j .
- the implicit links search enhancement system and method constructs an implicit link graph instead of the traditional explicit link graph in a small web sub-space.
- each implicit link I i,j ⁇ E′ is associated with a new parameter P(w j
- the implicit links search enhancement system and method disclosed herein extracts implicit links E′ by analyzing the observed users' browsing behaviors contained in a user access log.
- E′ controls how the user traverses in the small web.
- G′ and explicit link graph G, it can be assumed that there exists a generative model for the user access log.
- the model controls the generation of the user access log based on implicit links and explicit links.
- the final user access log contains abundant information on all implicit links.
- implicit links can be extracted by analyzing the observed explicit paths in the user access log.
- the implicit links search enhancement system 100 contains a number of modules. The operational details of these modules now will be discussed.
- FIG. 6 is a detailed flow diagram illustrating the operation of the user access log preprocessing module 300 shown in FIG. 3 .
- the user access log preprocessing module 300 initially inputs a user access log (box 600 ) and then performs data cleaning on the log (box 610 ). Data cleaning is done by filtering out any access entries for embedded objects, such as images and scripts.
- session identification is performed (box 620 ). All users are distinguished by their IP address. This assumes that consecutive accesses from the same IP address during a certain time interval are from a same user.
- consecutive repetition elimination is performed (box 630 ). This elimination handles the case of multiple users that have the same IP address. In particular, IP addresses whose page hits count exceeds some threshold are removed. The consecutive entries are then grouped into a browsing session. Different grouping criteria have been modeled and compared, as set forth in a paper by R. Cooley, B. Mobasher and J. Srivastava entitled “Data preparation for mining World Wide Web browsing patterns” in Knowledge and Information Systems, 1(1):5-32, 1999. Finally, the processed user access log is sent as output (box 640 ).
- FIG. 7 is a detailed flow diagram illustrating the operation of the user access log segmentation module 310 shown in FIG. 3 .
- the processed user access log is received an input (box 700 ).
- u i is the user identification and p ij are the pages in a browsing path ordered by timestamp.
- the segmented user access log is sent as output (box 720 ).
- FIG. 8 is a detailed flow diagram illustrating the operation of the ordered pairs generator 320 shown in FIG. 3 .
- the ordered pairs generator 320 uses a two-item sequential pattern mining technique to discover (or generate) possible implicit links. This technique uses a gliding window to move over each explicit path, generating all the ordered pairs and counting the occurrence of each distinct pair.
- the gliding window size represents the maximum interval a user clicks between the source page and the target page. For example, for an explicit path (w i1 , w i2 , w i3 , . . . , w ik ), the technique generates pairs (i 1 , i 2 ), (i 1 , i 3 ), . . .
- the individual browsing session from the segmented user access log are received as input (box 800 ).
- a gliding window size is defined (box 810 ).
- the gliding window is used to move over the path within each session to generate ordered pairs of pages.
- the gliding window size represents the maximum intervals users click between a source page and a target page.
- the gliding window then is applied to each individual browsing session (box 820 ).
- all possible ordered pairs are generated from each of the individual browsing sessions (box 830 ).
- the order pairs then are sent as output (box 840 ).
- FIG. 9 is a detailed flow diagram illustrating the operation of the filter module 330 shown in FIG. 3 .
- All possible ordered pairs and their frequency are calculated from all the browsing sessions S, and infrequent occurrences are filtered by a minimum support threshold.
- supp(i) refers to the percentage of the sessions that contain the item i.
- the support of a two-item pair (i, j), denoted supp(i, j), is defined in a similar way.
- a two-item ordered pair is frequent if its support supp(i, j) ⁇ min-supp, where min_supp is a user specified number.
- the ordered pairs are receive as input (box 900 ) and the frequency of each of the ordered pairs is determined (box 910 ).
- the minimum support threshold is defined (box 920 ) and applied to the frequency of each of the order pairs (box 930 ).
- a determination then is made whether the frequency is above the threshold (box 940 ). If not, then the ordered pair is discarded (box 950 ). Otherwise, the ordered pair is kept (box 960 ).
- the filtered two-item sequential patterns then are sent as output (box 970 ).
- FIG. 10 is a detailed flow diagram illustrating the operation of the re-ranking module 350 shown in FIG. 3 .
- the re-ranking module 350 inputs the updated implicit link graph or structure (box 1000 ).
- an adjacency matrix is defined to describe the implicit link graph (box 1010 ).
- a modified re-ranking formula is defined in terms of the adjacency matrix (box 1020 ).
- Search results are re-ranked using a modified link analysis technique (box 1030 ).
- the modified link analysis technique includes using the modified re-ranking formula and at least one type of re-ranking technique.
- One type of re-ranking technique is a score based re-ranking technique.
- Another type of re-ranking technique is an order based re-ranking technique.
- the order-based re-ranking technique is used.
- the re-ranked search results then are sent as output (box 1040 ).
- a modified link analysis technique is used to re-rank the search results obtained from a traditional search engine.
- the modified link analysis technique is based on the PageRank link analysis algorithm that is modified with novel modifications.
- the traditional PageRank algorithm is described in a paper by L. Page et al. entitled “The PageRank citation ranking: bringing order to the Web”.
- the modified PageRank links analysis technique works as follows. First, an adjacency matrix is constructed to describe the implicit links graph. In particular, assume the graph contains n pages. The n ⁇ n adjacency matrix is denoted by A and the entries A[i, j] is defined to be the weight of the implicit links I i,j . The adjacency matrix is used to compute the rank score of each page. In an “ideal” form, the rank score PR i of page w i is evaluated by a function on the rank scores of all the pages that point to page w i :
- the basic model is modified to obtain an “actual model” using a random walk technique.
- a random walk technique is used to modify the ranking formula to the following form:
- PR ⁇ ⁇ n ⁇ ⁇ e ⁇ + ( 1 - ⁇ ) ⁇ ⁇ A ⁇ ⁇ PR ⁇
- ⁇ right arrow over (e) ⁇ is the vector of all 1's
- ⁇ (0 ⁇ 1) is the probability parameter.
- the probability parameter ⁇ is set to 0.15.
- the modified links analysis technique also uses at least one type of re-ranking technique: (1) a score based re-ranking technique; and (2) an order based re-ranking technique.
- Order based re-ranking is based on the rank orders of the web-pages.
- the implicit links search enhancement system and method disclosed herein improves local searches (such as performed on a web site or intranet) by analyzing a user's access pattern by mining a user access log.
- the web site a having 4-month click-thru logs was used.
- the log was preprocessed by performing data cleaning, session identification and consecutive repetition elimination. Data cleaning was performed by filtering out the access entries for embedded objects such as images and scripts. Afterward, users were distinguished by their IP address. In other words, it was assumed that consecutive accesses from the same IP during a certain time interval were from a same user.
- IP addresses whose page hits count exceeds some threshold were removed.
- the consecutive entries then were grouped into a browsing session. Different grouping criteria were modeled and compared. This is detailed in a paper referenced above by Cooley et al. entitled “Data preparation for mining World Wide Web browsing patterns”.
- the “overtime” criterion was selected. More specifically, a new session starts when the duration of the whole group of traversals exceeds a time threshold. Consecutive repetitions within a session then are eliminated. For example, session (A, A, B, C, C, C) is compacted to (A, B, C). After preprocessing, the log contained about 300,000 transactions, 170,000 pages and 60,000 users.
- the window size was set at 4
- the minimum support threshold was set at 7
- a support-weighted adjacent matrix was used, and an order-based re-ranking technique was used for search.
- the implicit links search enhancement system and method was compared with several state-of-the-art algorithms including full text search, explicit link-based PageRank, DirectHit, and modified-HITS algorithm.
- 336,812 implicit links are generated. There are 22,122 links that are both in the explicit links and the generated implicit links. In other words, 11% of the links are overlapped. this is a relatively small number of overlapping links.
- Prediction ⁇ ⁇ precision P + ( P + + P - ) where P+ and P ⁇ are the numbers of correct and incorrect predictions, respectively.
- FIG. 11 illustrates the precision of page prediction by implicit links in the working example. As shown in FIG. 11 , the prediction precision monotonously increases as the minimum support increases. This indicates that the implicit links of the implicit links web search engine and method are accurate and reflects user's behaviors and interests.
- the quality of implicit links is evaluated from human perspective. Three subsets were randomly selected that contain 375 implicit links in total. Seven volunteer graduate students who are familiar with the subjects of the pages were chosen as evaluation subjects. They are asked to evaluate whether the implicit links are recommendation links according to the content of the pages. As shown in the upper part of Table 1, about 67% of implicit links in average are recommendation links. Another three subsets selected from explicit links are shown in the lower part of Table 1. Here, the average recommendation link ratio is about 39%.
- the precision measures the degree to which the algorithm produces an accurate result; while the authority measures the ability of the algorithm to produce the pages that are most likely to be visited by the user or the authority measurement is more relevant to user's satisfactory degree on the performance of a local (or small web) search engine.
- FIG. 12 is a bar graph illustrating the precision and authority of the different ranking methods.
- iPR denotes the implicit links search enhancement system and method (or implicit link-based PageRank)
- ePR, mH and DH correspond to explicit link-based PageRank, modified-HITS, and DirectHit, respectively.
- the right-most label “Avg” stands for the average value for the 10 queries.
- the implicit links web search engine and method outperforms the other 4 algorithms.
- the average improvement of precision over the full text is 16%, PageRank 20%, DirectHit 14% and modified-HITS 12%.
- the average improvement of authority over the full text is 24%, PageRank 26%, DirectHit 15% and modified-HITS 14%. From FIG. 12 , it can also been seen that the performance of explicit link-based PageRank is even worse than that of the full text search technique, demonstrating the unreliability of explicit link structure of this website.
- DirectHit has a medium performance in all the algorithms. DirectHit outperforms full text search because it takes usage information into account. However, DirectHit could not reveal the real authoritativeness of web-pages. The experiment also shows that DirectHit only improves a part of popular queries' precision. Thus, the average precision is not as good as the implicit links search enhancement system and method.
- the modified-HITS algorithm achieves higher performance than full text search, DirectHit and explicit link-based PageRank.
- this algorithm is a special case of the implicit links search enhancement system and method when the minimum support threshold is set to 0 and window size is set to 1. However, as mentioned above, when the minimum support threshold is set to 0, a great deal of noise data will be created. When the window size is set to 1, many useful links will be missed and this also affects performance.
- Table 3 shows the top 10 pages for the query “vision.” It was also found that the results from implicit link-based PageRank are more authoritative than that from the modified-HITS.
- “ANSI Common Lisp” is a page ranked high by explicit link-based PageRank because contains numerous out-links and in-links. But according the user logs, this page is rarely accessed.
- FIG. 13 illustrates the convergence curves of different ranking models.
- the gap represents the difference of the sum of page scores from previous iterations.
- the difference of PageRank values between consecutive iterations drops significantly after 7 iterations and shows a strong tendency toward zero. This illustrates the convergence of the implicit links web search engine and method in a practical way.
- FIGS. 14A and 14B illustrate the search precision and implicit link number with different minimum support thresholds.
- the implicit links search enhancement system and method achieved the best search precision when the minimum support is 7.
- FIG. 14A also illustrates that the system performance dramatically drops when the minimum support threshold is less than 4 or higher than 10. From these observations, the reason can be explained as follows. First, when the minimum support threshold is too small, user's random behaviors are counted and the number of the implicit links is large. This is illustrated in FIG.
- FIG. 15 illustrates the impact of different window sizes on search precision.
- the evaluation method is same as above. From FIG. 15 , it was found that the precision increases when the window size changes from 1 to 4. This proves the analysis set forth in a paper by P. Berkhin, J. D. Becher, and D. J. Randall entitled “Interactive path analysis of web site traffic” in Proc. of the 7 th SIGKDD, 414-410, San Francisco, Calif., 2001, that a user may click several times to get what is desired. On the other hand, by analyzing the effect of window size on the number of implicit links, it was found that more noisy implicit links are created if the window size is large.
- FIG. 16 illustrates an interval distribution of implicit links. As shown in FIG. 16 , about 13.7% of the implicit link is accessed in one step, 26% in two steps, 24% in three steps, and so on.
- FIG. 17 illustrates the precision of different weighting methods.
- the support-weighted method achieves better search precision compared to 0-1 weighted method in average. The improvement may be due to the fact that the support-weighted method has stronger recommendation than the 0-1 weighted.
- FIG. 18 illustrates the precision of various re-ranking techniques.
- the order-based re-ranking outperforms the score-based re-ranking.
- the probabilities could be easily calculated as in Table 1, where w x1 , w x2 , . . . ⁇ w i , w j and w x1 , w x2 , . . . are different from each other.
- This probability is calculated given current implicit link I i,j whose probability is P(w j
- this joint probability is the contribution of implicit link I i,j to the probability of explicit link I x,y .
- the contributions of all the implicit links can be summed to get the total probability of this explicit link P(w y
- the contribution of implicit links is ignored, with one end being w x or w y because
- the average probability of an implicit link I i,j is P(w j
- the average probability of explicit links is about (1 ⁇ p) 2 of that of implicit links.
- implicit links could be separated from explicit links by setting an appropriate minimum support.
- the variance of implicit link probabilities are relatively larger than the variance of explicit link probabilities (i.e., the users have no significant bias in selecting paths), most two-item access patterns obtained from web log mining with the highest support values will be implicit links.
Abstract
Description
The final authority and hub scores of every web page are obtained through an iterative update process.
where ε is a dampening factor, which is usually set between 0.1 and 0.2, n is the number of nodes in G, and out-degree(wj) is the number of the edges leaving page wj (i.e., the number of hyperlinks on page wj). The PageRank can be computed by an iterative algorithm and corresponds to the primary eigenvector of a matrix derived from adjacency matrix of the available portion of the Web.
-
- (1) Randomly select a page wi from V as the starting point;
- (2) Generate an implicit path (wi, wj, wk, . . . ) according to the implicit links E′ and the associated probabilities, where it is assumed each selection of implicit link is independent on previous selections;
- (3) For each pair of adjacent pages wi and wj in the implicit path, randomly select a set of in-between pages wx1, wx2, . . . , wxm according to the explicit links E to form an explicit path (wi, wx1, wx2, . . . , wxm, wj).
This recursive definition gives each page a fraction of the rank of each page pointing to it—inversely weighted by the strength of the links of that page. The above equation can be written in the form of matrix as:
{right arrow over (PR)}={right arrow over (APR)}
Or, written in matrix form:
where {right arrow over (e)} is the vector of all 1's, and ε(0<ε<1) is the probability parameter. In a preferred embodiment, the probability parameter ε is set to 0.15. Instead of computing an eigenvector, a Jacobi iteration iterative method is used to resolve the equation.
Score(w)=αSim+(1−α)PR(αε[0, 1])
where Sim is the content-based similarity between web-pages and query words, and PR is the PageRank value.
Score(w)=αO Sim+(1−α)O PR(αε[0, 1])
where OSim and OPR are positions of page w in similarity score list and PageRank list, respectively.
VII. Working Example
where P+ and P− are the numbers of correct and incorrect predictions, respectively.
TABLE 1 |
Recommendation links in implicit and explicit links. |
Subset | Recomm. link | Ratio | |||
|
|||||
1 | 128 | 87 | 0.68 | ||
2 | 114 | 82 | 0.72 | ||
3 | 133 | 84 | 0.63 |
Average | 0.67 |
|
||||
1 | 107 | 47 | 0.44 | |
2 | 84 | 26 | 0.31 | |
3 | 99 | 42 | 0.42 |
Average | 0.39 | ||
TABLE 2 |
Examples of the implicit links. |
Exp. | ||||
# | Source Page | Target Page | Explanation | Link? |
1. | Book: Artificial | The book's | A book and | Y |
Intelligence: A | slide | its slides | ||
| ||||
Approach | ||||
2. | Jordan's | Andrew Ng's | Teacher | Y |
Homepage | Homepage | and | ||
| ||||
3. | Various | Landscape | Picture | N |
pictures | photographs | |||
4. | Xuanlong's | Wilensky's | Same | N |
course: CS188 | course: | course | ||
5. | Anthony | Brian Harvey's | People in | N |
Joseph's | HomePage | | ||
Homepage | group | |||
6. | Al on the Web | Machine | Machine | N |
learning | | |||
software | ||||
7. | Sequin's | Sequin's | Course of | N |
course: CS284 | course: CS285 | same | ||
person | ||||
Search Result
TABLE 3 |
Ranks of query “vision” in different method. |
Web-page Descriptions | iPR | ePR | mH | DH | ||
|
1 | 41 | 2 | 8 | ||
Vision Group | ||||||
David Forsyth's Book: | 2 | 94 | 1 | 4 | ||
Computer Vision | ||||||
David Forsyth's Book: | 3 | 9 | 3 | 10 | ||
Computer Vision(3rd Draft) | ||||||
A workshop on Vision and | 4 | 44 | 20 | 1 | ||
Graphics | ||||||
|
5 | 2 | 13 | 7 | ||
| ||||||
CS | ||||||
280 |
6 | 14 | 10 | 11 | ||
Thomas Leung's | 7 | 55 | 4 | 31 | ||
Publications | ||||||
Jitendra Malik's |
8 | 17 | 7 | 6 | ||
Biography | ||||||
An overview of Grouping and | 9 | 5 | 21 | 5 | ||
Perceptual Organization | ||||||
David Forsyth's |
10 | 87 | 29 | 35 | ||
A paper of Phil | 13 | 1 | 6 | 9 | ||
Kim' ZuWhan resume | 18 | 3 | 5 | 2 | ||
A slide of Landay's talk about | 37 | 4 | 33 | 14 | ||
Notepals | ||||||
John A. Haddon's publication | 39 | 23 | 41 | 13 | ||
A slide of Landay's talk about | 41 | 6 | 42 | 18 | ||
Notepals | ||||||
Chris Bregler's Publications | 44 | 27 | 8 | 42 | ||
Course: Appearance Models | 51 | 63 | 47 | 3 | ||
for Computer Graphics and | ||||||
Vision | ||||||
Reference of Object | 62 | 59 | 9 | 17 | ||
Recognition | ||||||
-
- (a) Select the target page wj with probability p (0<p<1); select another page wx≠wj, wi with
probability 1−p. - (b) If arrived at wj, stop; else go to (a).
- (a) Select the target page wj with probability p (0<p<1); select another page wx≠wj, wi with
Claims (19)
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/676,794 US7584181B2 (en) | 2003-09-30 | 2003-09-30 | Implicit links search enhancement system and method for search engines using implicit links generated by mining user access patterns |
US12/505,426 US8312035B2 (en) | 2003-09-30 | 2009-07-17 | Search engine enhancement using mined implicit links |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/676,794 US7584181B2 (en) | 2003-09-30 | 2003-09-30 | Implicit links search enhancement system and method for search engines using implicit links generated by mining user access patterns |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US12/505,426 Division US8312035B2 (en) | 2003-09-30 | 2009-07-17 | Search engine enhancement using mined implicit links |
Publications (2)
Publication Number | Publication Date |
---|---|
US20050071465A1 US20050071465A1 (en) | 2005-03-31 |
US7584181B2 true US7584181B2 (en) | 2009-09-01 |
Family
ID=34377455
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/676,794 Expired - Fee Related US7584181B2 (en) | 2003-09-30 | 2003-09-30 | Implicit links search enhancement system and method for search engines using implicit links generated by mining user access patterns |
US12/505,426 Expired - Fee Related US8312035B2 (en) | 2003-09-30 | 2009-07-17 | Search engine enhancement using mined implicit links |
Family Applications After (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US12/505,426 Expired - Fee Related US8312035B2 (en) | 2003-09-30 | 2009-07-17 | Search engine enhancement using mined implicit links |
Country Status (1)
Country | Link |
---|---|
US (2) | US7584181B2 (en) |
Cited By (36)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070050708A1 (en) * | 2005-03-30 | 2007-03-01 | Suhit Gupta | Systems and methods for content extraction |
US20070271270A1 (en) * | 2006-05-19 | 2007-11-22 | Yahoo! Inc. | System and method for selecting object metadata evolving over time |
US20080092032A1 (en) * | 2003-10-16 | 2008-04-17 | International Business Machines Corporation | Web page development environment that displays frequency of use information |
US20090248661A1 (en) * | 2008-03-28 | 2009-10-01 | Microsoft Corporation | Identifying relevant information sources from user activity |
US20090292673A1 (en) * | 2008-05-22 | 2009-11-26 | Carroll Martin D | Electronic Document Processing with Automatic Generation of Links to Cited References |
US7716225B1 (en) * | 2004-06-17 | 2010-05-11 | Google Inc. | Ranking documents based on user behavior and/or feature data |
US20100251210A1 (en) * | 2009-03-24 | 2010-09-30 | International Business Machines Corporation | Mining sequential patterns in weighted directed graphs |
US8375073B1 (en) * | 2007-11-12 | 2013-02-12 | Google Inc. | Identification and ranking of news stories of interest |
US8380705B2 (en) | 2003-09-12 | 2013-02-19 | Google Inc. | Methods and systems for improving a search ranking using related queries |
US8396865B1 (en) | 2008-12-10 | 2013-03-12 | Google Inc. | Sharing search engine relevance data between corpora |
WO2013103694A1 (en) * | 2012-01-03 | 2013-07-11 | Microsoft Corporation | Search engine performance evaluation using a task-based assessment metric |
US8498974B1 (en) | 2009-08-31 | 2013-07-30 | Google Inc. | Refining search results |
US8615514B1 (en) | 2010-02-03 | 2013-12-24 | Google Inc. | Evaluating website properties by partitioning user feedback |
US20130346447A1 (en) * | 2012-06-21 | 2013-12-26 | Xerox Corporation | Systems and methods for behavioral pattern mining |
US8661029B1 (en) | 2006-11-02 | 2014-02-25 | Google Inc. | Modifying search result ranking based on implicit user feedback |
US8694511B1 (en) | 2007-08-20 | 2014-04-08 | Google Inc. | Modifying search result ranking based on populations |
US8694374B1 (en) | 2007-03-14 | 2014-04-08 | Google Inc. | Detecting click spam |
US20140244613A1 (en) * | 2009-11-13 | 2014-08-28 | Oracle International Corporation | Method And System for Enterprise Search Navigation |
US8832083B1 (en) | 2010-07-23 | 2014-09-09 | Google Inc. | Combining user feedback |
US8874555B1 (en) | 2009-11-20 | 2014-10-28 | Google Inc. | Modifying scoring data based on historical changes |
US8909655B1 (en) | 2007-10-11 | 2014-12-09 | Google Inc. | Time based ranking |
US8924379B1 (en) | 2010-03-05 | 2014-12-30 | Google Inc. | Temporal-based score adjustments |
US8938463B1 (en) | 2007-03-12 | 2015-01-20 | Google Inc. | Modifying search result ranking based on implicit user feedback and a model of presentation bias |
US8959093B1 (en) | 2010-03-15 | 2015-02-17 | Google Inc. | Ranking search results based on anchors |
US8972391B1 (en) | 2009-10-02 | 2015-03-03 | Google Inc. | Recent interest based relevance scoring |
US8972394B1 (en) | 2009-07-20 | 2015-03-03 | Google Inc. | Generating a related set of documents for an initial set of documents |
US9002867B1 (en) | 2010-12-30 | 2015-04-07 | Google Inc. | Modifying ranking data based on document changes |
US9009146B1 (en) | 2009-04-08 | 2015-04-14 | Google Inc. | Ranking search results based on similar queries |
US9092510B1 (en) | 2007-04-30 | 2015-07-28 | Google Inc. | Modifying search result ranking based on a temporal element of user feedback |
US9183499B1 (en) | 2013-04-19 | 2015-11-10 | Google Inc. | Evaluating quality based on neighbor features |
US9623119B1 (en) | 2010-06-29 | 2017-04-18 | Google Inc. | Accentuating search results |
US9639627B2 (en) | 2010-07-26 | 2017-05-02 | Hewlett-Packard Development Company, L.P. | Method to search a task-based web interaction |
US20170132327A1 (en) * | 2011-05-20 | 2017-05-11 | Amazon Technologies, Inc. | Pre-fetching of network page content in a browser application |
US10789417B1 (en) * | 2012-05-24 | 2020-09-29 | The Boeing Company | Method and apparatus for identifying relationships between documents |
US11366872B1 (en) * | 2017-07-19 | 2022-06-21 | Amazon Technologies, Inc. | Digital navigation menus with dynamic content placement |
US11397745B1 (en) | 2019-03-26 | 2022-07-26 | Grant Carter Hemingway | System and method for determining rankings, searching, and generating reports of profiles and personal information |
Families Citing this family (76)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7398271B1 (en) * | 2001-04-16 | 2008-07-08 | Yahoo! Inc. | Using network traffic logs for search enhancement |
WO2005029362A1 (en) * | 2003-09-22 | 2005-03-31 | Eurekster, Inc. | Enhanced search engine |
US8521725B1 (en) | 2003-12-03 | 2013-08-27 | Google Inc. | Systems and methods for improved searching |
US7464075B2 (en) * | 2004-01-05 | 2008-12-09 | Microsoft Corporation | Personalization of web page search rankings |
US7257577B2 (en) * | 2004-05-07 | 2007-08-14 | International Business Machines Corporation | System, method and service for ranking search results using a modular scoring system |
JP2006065395A (en) * | 2004-08-24 | 2006-03-09 | Fujitsu Ltd | Hyper link generating device, hyper link generating method, and hyper link generating program |
US7694311B2 (en) * | 2004-09-29 | 2010-04-06 | International Business Machines Corporation | Grammar-based task analysis of web logs |
US8056128B1 (en) * | 2004-09-30 | 2011-11-08 | Google Inc. | Systems and methods for detecting potential communications fraud |
US7539668B2 (en) * | 2004-11-30 | 2009-05-26 | International Business Machines Corporation | System and method for sorting data records contained in a query result based on suitability score |
US8538969B2 (en) * | 2005-06-03 | 2013-09-17 | Adobe Systems Incorporated | Data format for website traffic statistics |
JP2007140973A (en) * | 2005-11-18 | 2007-06-07 | National Institute Of Information & Communication Technology | Page reranking device, and page reranking program |
US7580931B2 (en) * | 2006-03-13 | 2009-08-25 | Microsoft Corporation | Topic distillation via subsite retrieval |
US20070239768A1 (en) * | 2006-04-10 | 2007-10-11 | Graphwise Llc | System and method for creating a dynamic database for use in graphical representations of tabular data |
US20070250855A1 (en) * | 2006-04-10 | 2007-10-25 | Graphwise, Llc | Search engine for presenting to a user a display having both graphed search results and selected advertisements |
US20070239698A1 (en) * | 2006-04-10 | 2007-10-11 | Graphwise, Llc | Search engine for evaluating queries from a user and presenting to the user graphed search results |
US20070240050A1 (en) * | 2006-04-10 | 2007-10-11 | Graphwise, Llc | System and method for presenting to a user a preferred graphical representation of tabular data |
US20070239686A1 (en) * | 2006-04-11 | 2007-10-11 | Graphwise, Llc | Search engine for presenting to a user a display having graphed search results presented as thumbnail presentations |
US8131703B2 (en) * | 2006-04-14 | 2012-03-06 | Adobe Systems Incorporated | Analytics based generation of ordered lists, search engine feed data, and sitemaps |
US20070255687A1 (en) * | 2006-04-27 | 2007-11-01 | Al-Yousuf Ahmed K | Research report search system |
US20070266025A1 (en) * | 2006-05-12 | 2007-11-15 | Microsoft Corporation | Implicit tokenized result ranking |
US7631263B2 (en) * | 2006-06-02 | 2009-12-08 | Scenera Technologies, Llc | Methods, systems, and computer program products for characterizing links to resources not activated |
CN100596135C (en) * | 2006-06-09 | 2010-03-24 | 华为技术有限公司 | System and method for determining content supplier priority level |
US7634476B2 (en) * | 2006-07-25 | 2009-12-15 | Microsoft Corporation | Ranking of web sites by aggregating web page ranks |
US20080071797A1 (en) * | 2006-09-15 | 2008-03-20 | Thornton Nathaniel L | System and method to calculate average link growth on search engines for a keyword |
US9646089B2 (en) * | 2006-09-18 | 2017-05-09 | John Nicholas and Kristin Gross Trust | System and method of modifying ranking for internet accessible documents |
US20080114753A1 (en) * | 2006-11-15 | 2008-05-15 | Apmath Ltd. | Method and a device for ranking linked documents |
US7617194B2 (en) * | 2006-12-29 | 2009-11-10 | Microsoft Corporation | Supervised ranking of vertices of a directed graph |
ITBG20070012A1 (en) * | 2007-02-13 | 2008-08-14 | Web Lion Sas | SEARCH METHOD AND SELECTION OF WEB SITES |
US7822755B2 (en) * | 2007-03-06 | 2010-10-26 | Yahoo! Inc. | Methods of processing and segmenting web usage information |
US7945572B2 (en) * | 2007-03-21 | 2011-05-17 | The Johns Hopkins University | Pattern discovery through reversing time flow |
US8996681B2 (en) * | 2007-04-23 | 2015-03-31 | The Mitre Corporation | Passively attributing anonymous network events to their associated users |
US7970721B2 (en) * | 2007-06-15 | 2011-06-28 | Microsoft Corporation | Learning and reasoning from web projections |
US8290986B2 (en) * | 2007-06-27 | 2012-10-16 | Yahoo! Inc. | Determining quality measures for web objects based on searcher behavior |
US8200694B1 (en) | 2007-07-23 | 2012-06-12 | Google Inc. | Identification of implicitly local queries |
US9178848B1 (en) | 2007-07-23 | 2015-11-03 | Google Inc. | Identifying affiliated domains |
US7783623B2 (en) * | 2007-08-31 | 2010-08-24 | Yahoo! Inc. | System and method for recommending songs |
US20090070168A1 (en) * | 2007-09-07 | 2009-03-12 | Power Measurement Ltd. | Enterprise energy management system with social network approach to data analysis |
US10115124B1 (en) * | 2007-10-01 | 2018-10-30 | Google Llc | Systems and methods for preserving privacy |
US7984000B2 (en) | 2007-10-31 | 2011-07-19 | Microsoft Corporation | Predicting and using search engine switching behavior |
US8255386B1 (en) * | 2008-01-30 | 2012-08-28 | Google Inc. | Selection of documents to place in search index |
US7937383B2 (en) * | 2008-02-01 | 2011-05-03 | Microsoft Corporation | Generating anonymous log entries |
US9128945B1 (en) | 2008-05-16 | 2015-09-08 | Google Inc. | Query augmentation |
US20090313286A1 (en) * | 2008-06-17 | 2009-12-17 | Microsoft Corporation | Generating training data from click logs |
US8788490B1 (en) | 2008-06-27 | 2014-07-22 | Google Inc. | Link based locale identification for domains and domain content |
US8368698B2 (en) * | 2008-09-24 | 2013-02-05 | Microsoft Corporation | Calculating a webpage importance from a web browsing graph |
US8103599B2 (en) * | 2008-09-25 | 2012-01-24 | Microsoft Corporation | Calculating web page importance based on web behavior model |
US8156120B2 (en) * | 2008-10-22 | 2012-04-10 | James Brady | Information retrieval using user-generated metadata |
US9053197B2 (en) * | 2008-11-26 | 2015-06-09 | Red Hat, Inc. | Suggesting websites |
US8335757B2 (en) * | 2009-01-26 | 2012-12-18 | Microsoft Corporation | Extracting patterns from sequential data |
US8489537B2 (en) * | 2009-01-26 | 2013-07-16 | Microsoft Corporation | Segmenting sequential data with a finite state machine |
US9495460B2 (en) * | 2009-05-27 | 2016-11-15 | Microsoft Technology Licensing, Llc | Merging search results |
US8150843B2 (en) | 2009-07-02 | 2012-04-03 | International Business Machines Corporation | Generating search results based on user feedback |
US20110022938A1 (en) * | 2009-07-23 | 2011-01-27 | Dennis Wilkinson | Apparatus, method and system for modifying pages |
US8423568B2 (en) * | 2009-09-16 | 2013-04-16 | Microsoft Corporation | Query classification using implicit labels |
GB2475473B (en) * | 2009-11-04 | 2015-10-21 | Nds Ltd | User request based content ranking |
US8150859B2 (en) * | 2010-02-05 | 2012-04-03 | Microsoft Corporation | Semantic table of contents for search results |
US8983989B2 (en) | 2010-02-05 | 2015-03-17 | Microsoft Technology Licensing, Llc | Contextual queries |
US8260664B2 (en) * | 2010-02-05 | 2012-09-04 | Microsoft Corporation | Semantic advertising selection from lateral concepts and topics |
US8903794B2 (en) | 2010-02-05 | 2014-12-02 | Microsoft Corporation | Generating and presenting lateral concepts |
US8849807B2 (en) | 2010-05-25 | 2014-09-30 | Mark F. McLellan | Active search results page ranking technology |
WO2012051583A1 (en) | 2010-10-15 | 2012-04-19 | Ness Computing, Inc. | A computer system and method for analyzing data sets and providing personalized recommendations |
US20120095862A1 (en) | 2010-10-15 | 2012-04-19 | Ness Computing, Inc. (a Delaware Corportaion) | Computer system and method for analyzing data sets and generating personalized recommendations |
US8489604B1 (en) | 2010-10-26 | 2013-07-16 | Google Inc. | Automated resource selection process evaluation |
US8346792B1 (en) | 2010-11-09 | 2013-01-01 | Google Inc. | Query generation using structural similarity between documents |
US8751487B2 (en) | 2011-02-28 | 2014-06-10 | International Business Machines Corporation | Generating a semantic graph relating information assets using feedback re-enforced search and navigation |
US9646110B2 (en) | 2011-02-28 | 2017-05-09 | International Business Machines Corporation | Managing information assets using feedback re-enforced search and navigation |
US20130024449A1 (en) * | 2011-07-20 | 2013-01-24 | Ness Computing, Inc. | Method and apparatus for allowing users to augment searches |
WO2013013093A1 (en) | 2011-07-20 | 2013-01-24 | Ness Computing, Inc. | Method and apparatus for quickly evaluating entities |
US9355095B2 (en) * | 2011-12-30 | 2016-05-31 | Microsoft Technology Licensing, Llc | Click noise characterization model |
US9449106B2 (en) * | 2013-03-08 | 2016-09-20 | Opentable, Inc. | Context-based queryless presentation of recommendations |
CN103593411A (en) * | 2013-10-23 | 2014-02-19 | 江苏大学 | Method for testing combination properties of evaluation indexes of search engines and testing device |
US10146775B2 (en) * | 2015-06-07 | 2018-12-04 | Apple Inc. | Apparatus, system and method for string disambiguation and entity ranking |
US9942252B1 (en) * | 2015-09-08 | 2018-04-10 | EMC IP Holding Co. LLC | Graph-based techniques for detecting coordinated network attacks |
CN107026831B (en) * | 2016-09-26 | 2020-03-10 | 阿里巴巴集团控股有限公司 | Identity recognition method and device |
US11036797B2 (en) * | 2017-10-12 | 2021-06-15 | Adtran, Inc. | Efficient storage and utilization of a hierarchical data set |
US11151089B2 (en) * | 2018-10-29 | 2021-10-19 | EMC IP Holding Company LLC | Compression of log data using pattern recognition |
-
2003
- 2003-09-30 US US10/676,794 patent/US7584181B2/en not_active Expired - Fee Related
-
2009
- 2009-07-17 US US12/505,426 patent/US8312035B2/en not_active Expired - Fee Related
Non-Patent Citations (16)
Title |
---|
Agrawal R. and Srikant R., "Mining sequential patterns" In Proc. of ICDE'95, 3-14 Taiwan, Mar. 1995. |
Berkhin P., Becher J. D. and Randall D. J., "Interactive path analysis of web site traffic" In Proc. of the 7th SIGKDD, 414-419, San Francisco, California, 2001. |
Brin S. and Page L., "The anatomy of a large-scale hypertextual web search engine" In Proc. of WWW7, 107-117, Brisbane, Australia, Apr. 1998. |
Cooley R., Mobasher B. and Srivastava J., "Data preparation for mining World Wide Web browsing patterns" in Knowledge and Information Systems, 1(1):5-32, 1999. |
Cooley R., Tan P.-N. and Srivastava J., "Discovery of interesting usage patterns from web data" In Proc. of WEBKDD'99, 163-182, Aug. 1999. |
Gui-Rong Xue et al., Implicit Link Analysis for Small Web Search, Aug. 1, 2003, ACM, 56-63. * |
Hawking D., Voorhees E., Bailey P. and Craswell N., "Overview of TREC-8 web track" In Proc. of TREC-8, 131-150, Gaithersburg MD, Nov. 1999. |
Kleinberg J. M., "Authoritative sources in a hyperlinked environment" Journal of the ACM, 46(5):604-632, 1999. |
Miller J. C., Rae G. and Schaefer F., "Modifications of Kleinberg's HITS algorithm using matrix exponentiation and web log records" In Proc. of SIGIR'01, 444-445, New Orleans, Sep. 2001. |
Nakayama T., Kato H. and Yamane Y., "Discovering the gap between web site designers' expectations and users' behavior" In Proc. of WWW9, Amsterdam, May 2000. |
Page L., Brin S., Motwani R. and Winograd T., "The PageRank citation ranking: bringing order to the Web" Technical report, Stanford University Database Group, 1998.; pp. 1-17. |
Pei J., Han J., Mortazavi-asl B., and Zhu H., "Mining access patterns efficiently from web logs" In Proc. of PAKDD'00, 396-407, Kyoto, Apr. 2000. |
Robertson S. E., Walker S., Beaulieu M. M., and Gatford M., Payne A. "1995. Okapi at TREC-4" In Proc. of TREC-4, 73-96, NIST Special Publication 500-236, Oct. 1996. |
Talagala et al., Usage patterns of a Web-based image collection, Mar. 15-18, 1999, IEEE, 203-214. * |
Yang Q., Zhang H. H. and Li T., "Mining web logs for prediction models in WWW caching and prefetching" In Proc. of KDD'01, 473-478, Aug. 2001. |
Yu et al., Improving pseudo-relevance feedback in web information retrieval using web page segmentation, May 2003, 11-18. * |
Cited By (71)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8452758B2 (en) | 2003-09-12 | 2013-05-28 | Google Inc. | Methods and systems for improving a search ranking using related queries |
US8380705B2 (en) | 2003-09-12 | 2013-02-19 | Google Inc. | Methods and systems for improving a search ranking using related queries |
US20080092032A1 (en) * | 2003-10-16 | 2008-04-17 | International Business Machines Corporation | Web page development environment that displays frequency of use information |
US8117209B1 (en) | 2004-06-17 | 2012-02-14 | Google Inc. | Ranking documents based on user behavior and/or feature data |
US9305099B1 (en) | 2004-06-17 | 2016-04-05 | Google Inc. | Ranking documents based on user behavior and/or feature data |
US7716225B1 (en) * | 2004-06-17 | 2010-05-11 | Google Inc. | Ranking documents based on user behavior and/or feature data |
US10152520B1 (en) | 2004-06-17 | 2018-12-11 | Google Llc | Ranking documents based on user behavior and/or feature data |
US10650087B2 (en) | 2005-03-30 | 2020-05-12 | The Trustees Of Columbia University In The City Of New York | Systems and methods for content extraction from a mark-up language text accessible at an internet domain |
US20170031883A1 (en) * | 2005-03-30 | 2017-02-02 | The Trustees Of Columbia University In The City Of New York | Systems and methods for content extraction from a mark-up language text accessible at an internet domain |
US20070050708A1 (en) * | 2005-03-30 | 2007-03-01 | Suhit Gupta | Systems and methods for content extraction |
US9372838B2 (en) | 2005-03-30 | 2016-06-21 | The Trustees Of Columbia University In The City Of New York | Systems and methods for content extraction from mark-up language text accessible at an internet domain |
US10061753B2 (en) * | 2005-03-30 | 2018-08-28 | The Trustees Of Columbia University In The City Of New York | Systems and methods for content extraction from a mark-up language text accessible at an internet domain |
US8468445B2 (en) * | 2005-03-30 | 2013-06-18 | The Trustees Of Columbia University In The City Of New York | Systems and methods for content extraction |
US7739275B2 (en) * | 2006-05-19 | 2010-06-15 | Yahoo! Inc. | System and method for selecting object metadata evolving over time |
US20070271270A1 (en) * | 2006-05-19 | 2007-11-22 | Yahoo! Inc. | System and method for selecting object metadata evolving over time |
US11188544B1 (en) | 2006-11-02 | 2021-11-30 | Google Llc | Modifying search result ranking based on implicit user feedback |
US10229166B1 (en) | 2006-11-02 | 2019-03-12 | Google Llc | Modifying search result ranking based on implicit user feedback |
US11816114B1 (en) | 2006-11-02 | 2023-11-14 | Google Llc | Modifying search result ranking based on implicit user feedback |
US8661029B1 (en) | 2006-11-02 | 2014-02-25 | Google Inc. | Modifying search result ranking based on implicit user feedback |
US9811566B1 (en) | 2006-11-02 | 2017-11-07 | Google Inc. | Modifying search result ranking based on implicit user feedback |
US9235627B1 (en) | 2006-11-02 | 2016-01-12 | Google Inc. | Modifying search result ranking based on implicit user feedback |
US8938463B1 (en) | 2007-03-12 | 2015-01-20 | Google Inc. | Modifying search result ranking based on implicit user feedback and a model of presentation bias |
US8694374B1 (en) | 2007-03-14 | 2014-04-08 | Google Inc. | Detecting click spam |
US9092510B1 (en) | 2007-04-30 | 2015-07-28 | Google Inc. | Modifying search result ranking based on a temporal element of user feedback |
US8694511B1 (en) | 2007-08-20 | 2014-04-08 | Google Inc. | Modifying search result ranking based on populations |
US9152678B1 (en) | 2007-10-11 | 2015-10-06 | Google Inc. | Time based ranking |
US8909655B1 (en) | 2007-10-11 | 2014-12-09 | Google Inc. | Time based ranking |
US8667037B1 (en) * | 2007-11-12 | 2014-03-04 | Google Inc. | Identification and ranking of news stories of interest |
US8375073B1 (en) * | 2007-11-12 | 2013-02-12 | Google Inc. | Identification and ranking of news stories of interest |
US20140143259A1 (en) * | 2007-11-12 | 2014-05-22 | Google Inc. | Identification and Ranking of News Stories of Interest |
US9405805B2 (en) * | 2007-11-12 | 2016-08-02 | Google Inc. | Identification and ranking of news stories of interest |
US20090248661A1 (en) * | 2008-03-28 | 2009-10-01 | Microsoft Corporation | Identifying relevant information sources from user activity |
US9239884B2 (en) * | 2008-05-22 | 2016-01-19 | Alcatel Lucent | Electronic document processing with automatic generation of links to cited references |
US20090292673A1 (en) * | 2008-05-22 | 2009-11-26 | Carroll Martin D | Electronic Document Processing with Automatic Generation of Links to Cited References |
US8898152B1 (en) | 2008-12-10 | 2014-11-25 | Google Inc. | Sharing search engine relevance data |
US8396865B1 (en) | 2008-12-10 | 2013-03-12 | Google Inc. | Sharing search engine relevance data between corpora |
US8689172B2 (en) * | 2009-03-24 | 2014-04-01 | International Business Machines Corporation | Mining sequential patterns in weighted directed graphs |
US20120197854A1 (en) * | 2009-03-24 | 2012-08-02 | International Business Machines Corporation | Mining sequential patterns in weighted directed graphs |
US20100251210A1 (en) * | 2009-03-24 | 2010-09-30 | International Business Machines Corporation | Mining sequential patterns in weighted directed graphs |
US8683423B2 (en) * | 2009-03-24 | 2014-03-25 | International Business Machines Corporation | Mining sequential patterns in weighted directed graphs |
US9009146B1 (en) | 2009-04-08 | 2015-04-14 | Google Inc. | Ranking search results based on similar queries |
US8972394B1 (en) | 2009-07-20 | 2015-03-03 | Google Inc. | Generating a related set of documents for an initial set of documents |
US8977612B1 (en) | 2009-07-20 | 2015-03-10 | Google Inc. | Generating a related set of documents for an initial set of documents |
US9697259B1 (en) | 2009-08-31 | 2017-07-04 | Google Inc. | Refining search results |
US8738596B1 (en) | 2009-08-31 | 2014-05-27 | Google Inc. | Refining search results |
US8498974B1 (en) | 2009-08-31 | 2013-07-30 | Google Inc. | Refining search results |
US9418104B1 (en) | 2009-08-31 | 2016-08-16 | Google Inc. | Refining search results |
US8972391B1 (en) | 2009-10-02 | 2015-03-03 | Google Inc. | Recent interest based relevance scoring |
US9390143B2 (en) | 2009-10-02 | 2016-07-12 | Google Inc. | Recent interest based relevance scoring |
US10795883B2 (en) | 2009-11-13 | 2020-10-06 | Oracle International Corporation | Method and system for enterprise search navigation |
US20140244613A1 (en) * | 2009-11-13 | 2014-08-28 | Oracle International Corporation | Method And System for Enterprise Search Navigation |
US9311409B2 (en) * | 2009-11-13 | 2016-04-12 | Oracle International Corporation | Method and system for enterprise search navigation |
US8898153B1 (en) | 2009-11-20 | 2014-11-25 | Google Inc. | Modifying scoring data based on historical changes |
US8874555B1 (en) | 2009-11-20 | 2014-10-28 | Google Inc. | Modifying scoring data based on historical changes |
US8615514B1 (en) | 2010-02-03 | 2013-12-24 | Google Inc. | Evaluating website properties by partitioning user feedback |
US8924379B1 (en) | 2010-03-05 | 2014-12-30 | Google Inc. | Temporal-based score adjustments |
US8959093B1 (en) | 2010-03-15 | 2015-02-17 | Google Inc. | Ranking search results based on anchors |
US9623119B1 (en) | 2010-06-29 | 2017-04-18 | Google Inc. | Accentuating search results |
US8832083B1 (en) | 2010-07-23 | 2014-09-09 | Google Inc. | Combining user feedback |
US9639627B2 (en) | 2010-07-26 | 2017-05-02 | Hewlett-Packard Development Company, L.P. | Method to search a task-based web interaction |
US9002867B1 (en) | 2010-12-30 | 2015-04-07 | Google Inc. | Modifying ranking data based on document changes |
US20170132327A1 (en) * | 2011-05-20 | 2017-05-11 | Amazon Technologies, Inc. | Pre-fetching of network page content in a browser application |
US9773069B2 (en) * | 2011-05-20 | 2017-09-26 | Amazon Technologies, Inc. | Pre-fetching of network page content in a browser application |
US8930339B2 (en) | 2012-01-03 | 2015-01-06 | Microsoft Corporation | Search engine performance evaluation using a task-based assessment metric |
WO2013103694A1 (en) * | 2012-01-03 | 2013-07-11 | Microsoft Corporation | Search engine performance evaluation using a task-based assessment metric |
US10789417B1 (en) * | 2012-05-24 | 2020-09-29 | The Boeing Company | Method and apparatus for identifying relationships between documents |
US9305104B2 (en) * | 2012-06-21 | 2016-04-05 | Xerox Corporation | Systems and methods for behavioral pattern mining |
US20130346447A1 (en) * | 2012-06-21 | 2013-12-26 | Xerox Corporation | Systems and methods for behavioral pattern mining |
US9183499B1 (en) | 2013-04-19 | 2015-11-10 | Google Inc. | Evaluating quality based on neighbor features |
US11366872B1 (en) * | 2017-07-19 | 2022-06-21 | Amazon Technologies, Inc. | Digital navigation menus with dynamic content placement |
US11397745B1 (en) | 2019-03-26 | 2022-07-26 | Grant Carter Hemingway | System and method for determining rankings, searching, and generating reports of profiles and personal information |
Also Published As
Publication number | Publication date |
---|---|
US20050071465A1 (en) | 2005-03-31 |
US20100023508A1 (en) | 2010-01-28 |
US8312035B2 (en) | 2012-11-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7584181B2 (en) | Implicit links search enhancement system and method for search engines using implicit links generated by mining user access patterns | |
Xue et al. | Implicit link analysis for small web search | |
US7194454B2 (en) | Method for organizing records of database search activity by topical relevance | |
Menczer | Complementing search engines with online web mining agents | |
US7779001B2 (en) | Web page ranking with hierarchical considerations | |
Xue et al. | Optimizing web search using web click-through data | |
Menczer et al. | Topical web crawlers: Evaluating adaptive algorithms | |
Chakrabarti et al. | The structure of broad topics on the web | |
Chakrabarti et al. | Distributed hypertext resource discovery through examples | |
US7945565B2 (en) | Method and system for generating a hyperlink-click graph | |
US6490577B1 (en) | Search engine with user activity memory | |
Baeza-Yates | Applications of web query mining | |
Rungsawang et al. | Learnable topic-specific web crawler | |
US20070143279A1 (en) | Identifying important news reports from news home pages | |
US20080097958A1 (en) | Method and Apparatus for Retrieving and Indexing Hidden Pages | |
US20100082752A1 (en) | Query log mining for detecting spam hosts | |
Fu et al. | A Framework for Personal Web Usage Mining. | |
Dohare et al. | Novel web usage mining for web mining techniques | |
Lee et al. | An interactive agent-based system for concept-based web search | |
Agosti et al. | Information retrieval on the web | |
US20100082694A1 (en) | Query log mining for detecting spam-attracting queries | |
Dahiwale et al. | Design of improved focused web crawler by analyzing semantic nature of URL and anchor text | |
Menczer et al. | Topic-driven crawlers: Machine learning issues | |
Xue et al. | User Access Pattern Enhanced Small Web Search. | |
Almpanidis et al. | Focused crawling using latent semantic indexing–An application for vertical search engines |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: MICROSOFT CORPORATION, WASHINGTON Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:ZENG, HUA-JUN;XUE, GUI-RONG;CHEN, ZHENG;AND OTHERS;REEL/FRAME:022964/0971;SIGNING DATES FROM 20030927 TO 20031001 |
|
STCF | Information on status: patent grant |
Free format text: PATENTED CASE |
|
FPAY | Fee payment |
Year of fee payment: 4 |
|
AS | Assignment |
Owner name: MICROSOFT TECHNOLOGY LICENSING, LLC, WASHINGTON Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:MICROSOFT CORPORATION;REEL/FRAME:034541/0477 Effective date: 20141014 |
|
FPAY | Fee payment |
Year of fee payment: 8 |
|
FEPP | Fee payment procedure |
Free format text: MAINTENANCE FEE REMINDER MAILED (ORIGINAL EVENT CODE: REM.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
LAPS | Lapse for failure to pay maintenance fees |
Free format text: PATENT EXPIRED FOR FAILURE TO PAY MAINTENANCE FEES (ORIGINAL EVENT CODE: EXP.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
STCH | Information on status: patent discontinuation |
Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362 |
|
FP | Lapsed due to failure to pay maintenance fee |
Effective date: 20210901 |